CN105098869A - System and method for battery power management - Google Patents

Abstract

The invention relates to a system and method for battery power management. The battery management system includes a battery pack and a controller. The controller is configured to receive pack terminal voltage and current data. In response, the controller may estimate battery model parameters in an equivalent circuit model and output state variable values indicative of fast and slow dynamics of the voltage responses. The controller may also output parameter values indicative of feedback gains to compute a current limit in a state feedback structure. The controller may further estimate battery current limits based on the state variable values and the feedback gains to control operation of the battery pack.

Description

Method and system for battery power management

Technical Field

The present disclosure relates to battery management techniques that are capable of estimating parameters of a battery model to provide control of an associated battery.

Background

Hybrid Electric Vehicles (HEVs) utilize a combination of an internal combustion engine and an electric motor to provide power. This configuration provides improved fuel economy for a vehicle having only an internal combustion engine. One method of improving fuel economy in an HEV is to shut down the engine during periods when the engine is operating inefficiently and not otherwise needed to propel the vehicle. In these cases, the electric motor is used to provide all of the power required to propel the vehicle. When the driver demand for power increases such that the electric motor can no longer provide sufficient power to meet the demand, or in other situations such as a battery state of charge (SOC) drop to a certain level, the engine must start quickly and smoothly in a manner that is nearly transparent to the driver.

The HEV includes a battery management system that estimates values describing the current operating conditions of the battery pack and/or battery cells. Battery pack and/or battery cell operating conditions include: battery SOC, power fade, capacity fade, and instantaneous available power. The battery management system should be able to estimate the values during the entire life cycle of the battery pack as the battery cells age changing cell characteristics. Accurate estimation of some parameters may improve performance and robustness, and may ultimately extend the useful life of the battery pack.

Disclosure of Invention

A control system for a battery pack includes a controller that operates the battery pack within current limits derived from first and second state variable values and a feedback gain. The first state variable value is based on battery pack terminal voltage data and battery pack current data and has a frequency component greater than a threshold value. The second state variable value is based on battery pack terminal voltage data and battery pack current data and has a frequency component less than the threshold.

A battery management system, comprising: a battery pack; at least one controller configured to: operating the battery pack within current limits from a first state variable value and a second state variable value and a feedback gain, wherein the first state variable value is based on battery pack terminal voltage data and battery pack current data and has a frequency component greater than a threshold value, and the second state variable value is based on battery pack terminal voltage data and battery pack current data and has a frequency component less than the threshold value.

A method for controlling a battery pack, comprising: outputting a first state variable value and a second state variable value, wherein the first state variable value is based on battery terminal voltage data and battery current data and has a frequency component greater than a threshold value, and the second state variable value is based on battery terminal voltage data and battery current data and has a frequency component less than the threshold value; outputting a feedback gain associated with the first state variable and the second state variable based on impedance parameters defining an equivalent circuit model of the battery pack; outputting a current limit for the battery pack based on the feedback gain and the first state variable value and the second state variable value; operating the battery pack based on the limit.

According to one embodiment of the invention, the equivalent circuit model comprises two or more RC circuits.

According to one embodiment of the invention, the current limits comprise a discharge current limit and a charge current limit.

According to an embodiment of the invention, the method further comprises: operating the battery pack during a charging event or during a discharging event within power limits including an instantaneous power capacity derived from the battery pack terminal voltage data and the battery pack current data.

According to an embodiment of the invention, the method further comprises: operating the battery pack during a charging event or during a discharging event within power limits including a sustained power capacity derived from the battery pack terminal voltage data and the battery pack current data.

According to an embodiment of the invention, the method further comprises: operating the battery pack within a power limit having a predefined duration during a charging event or during a discharging event.

A control system for a battery pack includes a controller configured to: operating the battery pack within current limits derived from a first state variable based on the battery pack terminal voltage data and the battery pack current data and having a frequency component greater than a threshold value and a second state variable based on the battery pack terminal voltage data and the battery pack current data and having a frequency component less than the threshold value and a feedback gain.

According to one embodiment of the invention, the feedback gain is based on an impedance parameter of an equivalent circuit model defining the battery pack.

According to one embodiment of the invention, the equivalent circuit model comprises two or more RC circuits.

According to one embodiment of the invention, the current limits comprise a discharge current limit and a charge current limit.

According to an embodiment of the invention, the controller is further configured to: operating the battery pack during a charging event or during a discharging event within power limits including an instantaneous power capacity derived from the battery pack terminal voltage data and the battery pack current data.

According to an embodiment of the invention, the controller is further configured to: operating the battery pack during a charging event or during a discharging event within power limits including a sustained power capacity derived from the battery pack terminal voltage data and the battery pack current data.

According to an embodiment of the invention, the controller is further configured to: operating the battery pack within a power limit having a predefined duration during a charging event or during a discharging event.

Drawings

FIG. 1 is a schematic diagram of a hybrid electric vehicle illustrating a typical powertrain and energy storage assembly;

FIG. 2 is a graph of an electrical impedance spectroscopy Nyquist plot showing the impedance of the cell;

FIG. 3 is a schematic diagram of an equivalent circuit model using an RC circuit to model a battery;

FIG. 4 is a graph showing the frequency response of an equivalent circuit model with one RC circuit in a Nyquist plot;

FIG. 5 is a schematic diagram of an equivalent circuit model using two RC circuits to model a battery;

FIG. 6 is a graph showing the battery impedance calculated in the Nyquist plot using two RC circuits in the equivalent circuit model;

fig. 7 is a graph showing terminal voltage response of a battery according to values of state variables from different battery usage curves;

FIG. 8 is a generalized state feedback structure for estimating battery current limits;

FIG. 9 is a graph showing an estimated voltage response of a state variable in an equivalent circuit using two RC circuits during operation of a vehicle;

FIG. 10 is a graph showing current limits estimated by the proposed state feedback structure;

FIG. 11 is a flow chart of an algorithm for estimating battery current and power limits in a battery management system.

Detailed Description

Embodiments of the present disclosure are described herein. However, it is to be understood that the disclosed embodiments are merely exemplary, and that other embodiments may be embodied in various and alternative forms. The figures are not necessarily to scale, some features may be exaggerated or minimized to show details of particular components. Therefore, specific structural and functional details disclosed herein are not to be interpreted as limiting, but merely as a representative basis for teaching one skilled in the art to variously employ the embodiments. As one of ordinary skill in the art will appreciate, various features illustrated and described with reference to any one of the figures may be combined with features illustrated in one or more other figures to form embodiments that are not explicitly illustrated or described. The combination of features shown provides a representative embodiment for typical applications. However, various combinations and modifications of the features consistent with the teachings of the present disclosure may be desired for particular applications or implementations.

Embodiments of the present disclosure generally provide a plurality of circuits or other electrical devices. References to the circuits and other electrical devices and the functions provided by each of them are not intended to be limited to only encompassing what is shown and described herein. While particular reference numbers may be assigned to the various circuits or other electrical devices disclosed, such reference numbers are not intended to limit the operating range of the circuits and other electrical devices. The circuits and other electrical devices may be combined and/or separated from one another in any manner based on the particular type of electrical implementation desired. It should be appreciated that any circuit or other electrical device disclosed herein may include any number of microprocessors, integrated circuits, storage devices (e.g., flash memory, Random Access Memory (RAM), Read Only Memory (ROM), Electrically Programmable Read Only Memory (EPROM), Electrically Erasable Programmable Read Only Memory (EEPROM), or other suitable variations thereof) and software that cooperate with one another to perform the operations disclosed herein. Further, any one or more electrical devices may be configured to execute a computer program embodied in a non-transitory computer readable medium, where the computer program is written to perform any number of the disclosed functions.

An HEV battery system may implement a battery management strategy that estimates values describing the current operating conditions of the battery and/or one or more battery cells. The operating conditions of the battery pack and/or one or more battery cells include: battery state of charge, power fade, capacity fade, and instantaneous available power. The battery management strategy can estimate the value as the battery cells age throughout the life of the battery pack. Accurate estimation of some parameters may improve performance and robustness, and may ultimately extend the useful life of the battery pack. For the battery systems described herein, the estimation of parameters for some battery packs and/or cells may be accomplished as discussed below.

Fig. 1 shows a typical hybrid electric vehicle. A typical hybrid electric vehicle 2 may include one or more electric motors 4 mechanically connected to a hybrid transmission 6. Furthermore, the hybrid transmission 6 is mechanically connected to the engine 8. The hybrid transmission 6 is also mechanically connected to a drive shaft 10, wherein the drive shaft 10 is mechanically connected to the wheels 12. In another embodiment, not depicted in the figures, the hybrid transmission may be a non-selectable gear transmission, wherein the non-selectable gear transmission may include at least one electric motor. The electric motor 4 may provide propulsion and deceleration capabilities when the engine 8 is started or shut off. The electric motor 4 also acts as a generator and can provide fuel economy benefits by recovering energy that would normally be lost as heat in a friction braking system. The electric motor 4 may also provide reduced polluting emissions, since the hybrid electric vehicle 2 may be operated in an electric mode under certain conditions.

The battery pack 14 may include, but is not limited to, a traction battery having one or more battery cells that store energy that may be used by the electric motor 4. The vehicle battery pack 14 typically provides a high voltage DC output and is electrically connected to the power electronics module 16. The power electronics module 16 may communicate with one or more control modules that make up the vehicle computing system 22. The vehicle computing system 22 may control a number of vehicle functions, systems, and/or subsystems. The one or more modules may include, but are not limited to, a battery management system. The power electronics module 16 is also electrically connected to the electric motor 4 and provides the ability to transfer energy bi-directionally between the battery pack 14 and the electric motor 4. For example, a typical battery pack 14 may provide a DC voltage, while the electric motor 4 may require a three-phase AC current to operate. The power electronics module 16 may convert the DC voltage to three-phase AC current required by the electric motor 4. In the regeneration mode, the power electronics module 16 converts the three-phase AC current from the electric motor 4 acting as a generator to the DC voltage required by the battery pack 14.

In addition to providing energy for propulsion, the battery pack 14 may provide energy for other vehicle electrical systems. A typical system may include a DC/DC converter module 18, wherein the DC/DC converter module 18 converts the high voltage DC output of the battery pack 14 to a low voltage DC supply that is compatible with other vehicle loads. Other high voltage loads may be connected directly without the use of the DC/DC converter module 18. In a typical vehicle, the low voltage system is electrically connected to a 12V battery 20.

The battery pack 14 may be controlled by a power electronics module 16, and the power electronics module 16 may receive commands from a vehicle computing system 22 having one or more control modules. The one or more control modules may include, but are not limited to, a battery control module. The one or more control modules may be calibrated to control the battery pack 14 using a battery model parameter estimation method that estimates an average of the effective battery internal resistances during operation to determine the battery power capacity. Power capacity prediction enables battery pack 14 to prevent overcharging and overdischarging that may lead to reduced battery life, performance issues related to the vehicle driveline, and the like.

The battery parameter prediction method and/or strategy may facilitate determining battery current limits and power capacity in real-time (i.e., during operation). Many battery parameter estimation processes are affected by the fidelity of the battery model and unpredictable environmental conditions or unpredictable noise during battery operation. For example, if the battery is in a charge-depleting mode, a simple battery model may not be able to capture the complex system dynamics associated with the voltage output and current input that the battery model is attempting to measure. Vehicle battery measurement methods/strategies may use an equivalent circuit model that uses one or more resistance-capacitance (RC) circuits in several configurations to measure a battery pack within a vehicle during operation to obtain electrochemical impedance.

Calibration for controlling the battery pack may be accomplished using multiple tables to capture a wide range of frequencies that affect the impedance of the battery pack and its associated dynamic characteristics. In order to fill/calibrate the plurality of tables, off-line testing of the battery pack using a complex algorithm in the test equipment needs to be strictly performed. An example of an offline test battery is Electrochemical Impedance Spectroscopy (EIS), where EIS may be implemented to capture battery system characteristics over a wide range of frequencies, where the battery system characteristics may include battery temperature, battery state of charge, and/or battery usage.

The vehicle battery measurement method may be implemented to eliminate the need for extensive off-line testing. The vehicle battery measurement method may use one or more simple equivalent circuits to measure the battery pack within the vehicle during operation to obtain the electrochemical impedance. The on-board battery measurement estimation method may have a higher noise level than the off-line parameter estimation, however, the on-board battery parameter estimation method may provide valuable information about the battery transient behavior during vehicle operation.

The HEV battery management system may implement an equivalent circuit model for predicting battery performance, where battery parameters and estimated electrochemical impedance based on the next few seconds of battery measurement are used to predict battery performance. The estimated battery parameters may vary depending on driving conditions and electric vehicle operating modes, such as charge retention mode or charge consumption mode. Battery parameter estimation processes using simple equivalent circuit models tend to be sensitive to internal and external noise and environmental conditions.

The system may use the battery measurements to estimate battery model parameters, and then use the estimated model parameters to calculate the battery power capacity. The battery power capacity is affected by the impedance of the battery pack and its associated dynamics. The battery model parameter estimation method may include battery measurements within the vehicle to obtain electrochemical impedances using an extended kalman filter and other calculations/algorithms that will be described in more detail below to calculate battery power capacity. The power capacity of the battery may be determined by state variables and may be derived by using system inputs and outputs.

The model-based battery management system provides sufficient computational speed that can be managed in the battery management system based on the equivalent circuit model without introducing additional hardware and/or increasing system complexity. The characteristics of the battery system may be calculated via a real-time parameter estimation method on a battery model by using direct battery measurements within the HEV. The system may measure a battery current input and a battery terminal voltage. The measurements may be recorded, calculated, and stored in one or more control modules within a vehicle computing system that includes a battery energy control module.

Fig. 2 is a graph 100 illustrating an EIS nyquist plot of battery impedance versus frequency. The EIS nyquist diagram 100 shows a direct physical interpretation of a battery system using one equivalent circuit. The EIS nyquist plot 100 has an x-axis representing the real part of impedance 104 and a y-axis representing the imaginary part of impedance 102. Curve 106 shows the measured impedance of the cell over a range of frequency responses. The range of frequency response of the system may reveal the energy storage and consumption characteristics of the battery.

The EIS Nyquist plot 100 may reveal information about the electrochemical process of the cellInformation of reaction principles, wherein the reaction principles comprise different reaction steps that may dominate at a specific frequency, and the frequency response may help determine the rate limiting step. Curve 106 may represent a slow cell dynamic response caused by diffusion processes of solid particles of electrode active material and polarization processes throughout the cell thickness. Transient response is represented by the internal resistance term R of the equivalent circuit model of the battery₀110. The battery dynamics represented by the medium-high frequency 108 determine the power capacity mainly in consideration of the battery dynamics. In the equivalent circuit model, the slow dynamics represented by the low frequency 112 (e.g., the Walburg (Warburg) impedance term) and the dynamic represented by R₀The transient dynamics, denoted by 110, are modeled as real-time adjusted internal resistance in the equivalent circuit model. The graph 100 captures a battery dynamic response that may be used to estimate an instantaneous battery power capacity of the battery system.

Fig. 3 is a schematic diagram of an equivalent circuit with one RC circuit for modeling a battery. The circuit may model a battery that includes a battery pack and/or one or more battery cells. The equivalent circuit model includes an active electrolyte resistance (or internal resistance) R₀202. Parallel capacitor C₁204 and an active charge transfer resistance R₁206, wherein the active electrolyte resistance R₀202 and an active charge transfer resistor R connected in parallel₁206 and a capacitance C₁204 are connected in series. Battery dynamics and associated state variables are represented as terminal voltage output v_t212. Open circuit voltage v of battery_OC214. Voltage v of internal resistance₀216 and voltage v of the RC circuit₁210. The model may be implemented in an HEV battery management system to provide predictive calculations for one or more battery parameters.

Fig. 4 is a graph 301 showing the frequency response of an equivalent circuit model with one RC circuit in a nyquist plot. The x-axis 316 of the graph 301 represents the real part of the average battery impedance over a time window. The y-axis 314 of the graph 301 represents the imaginary part of the average electrical impedance of the battery cell. Medium fast dynamic characteristics are provided by RC circuits (i.e., R)₁And C₁) Resulting semi-closed loop (sem)i-circuit) 108', and internal resistance and R₀110' are related. However, the slow dynamics referred to as the Valley term 112' are not captured by an equivalent circuit model with an RC circuit. Thus, the slow dynamics known herein as the Valurburg term 112' cannot be effectively represented in the model of the one RC circuit.

Fig. 5 is a schematic diagram of a simple equivalent circuit model 400 that models a battery using two RC circuits, according to an embodiment. The two RC circuits may improve the modeling 400 of the battery pack and/or one or more battery cells by introducing additional dynamic characteristics to the model. For example, the slow dynamics 112 may be modeled using additional RC circuits. The RC circuit model may include an additional RC circuit, wherein the additional RC circuit has a resistor R₂406 and a capacitor C₂404, wherein the resistor R₂406 and a capacitor C₂Are connected in parallel with each other and in series with the RC circuit in the equivalent circuit model 200 shown in fig. 3. The equivalent circuit model may have other configurations not limited to one or two RC circuits. The equivalent circuit model may include, but is not limited to, two or more RC circuits to model the battery.

Fig. 6 is a graph 301' illustrating calculation of an average internal resistance of one or more battery cells using two or more RC circuits in an equivalent circuit model according to an embodiment. The x-axis 316 of the graph 301' represents the real part of the average battery impedance over a time window. The y-axis 314 of the graph 301' represents the imaginary component of the average electrical impedance for the battery cell.

Graph 301' shows that the system captures the average internal resistance in dependence on the high frequency 108 "as a component of the electrical impedance of one or more cells. The system may capture the low frequency 112 "component of the electrical impedance of the one or more cells by using two or more RC circuits in an equivalent circuit model. Under wide frequency range operation, the system can utilize improved fidelity to estimate battery current limit and power capacity (especially for vehicle operating conditions where slow dynamics become battery operated).

For example, medium fast dynamics are provided by RC circuits (i.e., R)₁And C₁) Resulting in a semi-closed loop 108 "and an internal resistance associated with R₀110 "are correlated. The slow dynamics, referred to as the Valley term 112 ", are modeled by an additional RC circuit (i.e., R) in the equivalent circuit model₂And C₂) And (4) capturing. Thus, the slow dynamics, known herein as the Valley term 112 ", are demonstrated in an equivalent circuit model using two or more RC circuits.

The vehicle battery measurement method may implement a simple equivalent circuit model 400 using two RC circuits to independently capture fast dynamics and slow dynamics. The two RC circuits may improve the predictive capability for low temperature and/or long duration charging conditions. The landles (Randles) circuit model 200 shown in fig. 3 is not able to capture the slow cell dynamics associated with the walbauer impedance term.

Two RC circuits can improve the modeling of the battery dynamics by capturing the low and medium high frequency responses using the following equations:

<math> <mrow> <msub> <mover> <mi>v</mi> <mo>&CenterDot;</mo> </mover> <mn>1</mn> </msub> <mo>=</mo> <mo>-</mo> <mfrac> <mn>1</mn> <mrow> <msub> <mi>R</mi> <mn>1</mn> </msub> <msub> <mi>C</mi> <mn>1</mn> </msub> </mrow> </mfrac> <msub> <mi>v</mi> <mn>1</mn> </msub> <mo>+</mo> <mfrac> <mn>1</mn> <msub> <mi>C</mi> <mn>1</mn> </msub> </mfrac> <mi>i</mi> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow> </math>

wherein v is₁210 is formed by a resistor R₁And a capacitor C₁Two RC circuitsTerminal voltage, resistance R₁206 is the active charge transfer resistance and i208 is the current that energizes the circuit. By a resistor R₁And a capacitor C₁The resulting RC circuit represents the battery dynamics during vehicle operation. By a resistor R₂And a capacitor C₂The composed RC circuit represents the slow dynamics (i.e., low frequency) of the battery during vehicle operation using the following equation:

<math> <mrow> <msub> <mover> <mi>v</mi> <mo>&CenterDot;</mo> </mover> <mn>2</mn> </msub> <mo>=</mo> <mo>-</mo> <mfrac> <mn>1</mn> <mrow> <msub> <mi>R</mi> <mn>2</mn> </msub> <msub> <mi>C</mi> <mn>2</mn> </msub> </mrow> </mfrac> <msub> <mi>v</mi> <mn>2</mn> </msub> <mo>+</mo> <mfrac> <mn>1</mn> <msub> <mi>C</mi> <mn>2</mn> </msub> </mfrac> <mi>i</mi> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </mrow> </math>

wherein v is₂408 is represented by R₂406 and C₂404, i208 is the current of the pump circuit. Having a resistor R₂406 and a capacitor C₂The additional RC circuit of (a) represents a low frequency during vehicle operation. An equivalent circuit model with two RC circuits can calculate the battery terminal voltage using the following equation:

v_t＝v_OC-v₁-v₂-R₀i(3)

wherein v is_t212 is terminal voltage, v_OC214 is the battery open circuit voltage, v₁210 is formed by a resistor R₁And a capacitor C₁Voltage across the constituent RC circuits, v₂408 is represented by R₂406 and C₂404 across the RC circuit₀202 is the internal resistance of the battery, the voltage across the RC circuit can be calculated using the following equation:

$v_1=v_{\mathrm{1,0}}{e}^{-\frac{1}{R_1{C}_1}t}+(1-e^{-\frac{1}{R_1{C}_1}t})R_{1}i---\left(4\right)$

$v_2=v_{\mathrm{2,0}}{e}^{-\frac{1}{R_2{C}_2}t}+(1-e^{-\frac{1}{R_2{C}_2}t})R_{2}i---\left(5\right)$

battery terminal voltage estimates using multiple RC equivalent circuit models are derived according to the following equation:

$v_t=v_{\mathrm{OC}}-v_{\mathrm{1,0}}{e}^{-\frac{1}{R_1{C}_1}t}-v_{\mathrm{2,0}}{e}^{-\frac{1}{R_2{C}_2}t}-(R_0+(1-e^{-\frac{1}{R_1{C}_1}t})R_1+(1-e^{-\frac{1}{R_2{C}_2}t})R_2)i---\left(6\right)$

where t is time.

The system may map the current time t according to the following equation_oThe battery terminal voltage response at (e.g., t equals 0) is linearized to obtain a generalized state feedback structure for estimating the battery current limit:

$v_t=v_t{|}_{t=0}+\frac{{\mathrm{dv}}_t}{\mathrm{dt}}{|}_{t=0}t---\left(7\right)$

wherein equation (7) is as follows:

$\frac{{\mathrm{dv}}_t}{\mathrm{dt}}{|}_{t=0}=\frac{v_{\mathrm{1,0}}}{R_1{C}_1}+\frac{v_{2,0}}{R_2{C}_2}-(\frac{1}{C_1}+\frac{1}{C_2})i---\left(8\right)$

wherein equation (8) represents the voltage change rate, and the voltage change rate is derived by setting t to 0 in the following equation:

$\frac{{\mathrm{dv}}_t}{\mathrm{dt}}=\frac{v_{\mathrm{1,0}}}{R_1{C}_1}{e}^{-\frac{1}{R_1{C}_1}t}+\frac{v_{\mathrm{2,0}}}{R_2{C}_2}{e}^{-\frac{1}{R_2{C}_2}t}-(\frac{1}{C_1}{e}^{-\frac{1}{R_1{C}_1}t}+\frac{1}{C_2}{e}^{-\frac{1}{R_2{C}_2}t})i---\left(9\right)$

FIG. 7 shows dependence on v in an equivalent circuit model_1,0And v_2,0The estimated terminal voltage of the battery state. When the battery input current is at v_1,0And v_2,0Are the same value, the voltage response with equal input current 508 is subject to equation (8)The direct influence of (c).

When voltage responsePositive, the rate of change of voltage moves in a positive direction 510. The voltage response with respect to time with the positive input current 510 lies without taking into account the battery internal dynamics (i.e., with v_1,00 and v_2,0Equal input current 508 of 0).

When voltage responseWhen negative, the rate of change of voltage moves in a negative direction 506. The voltage response with respect to time with negative input current 506 lies without regard to the dynamics of the battery interior (i.e., with v_1,00 and v_2,0Equal input current 508 of 0).

The duration Δ t may be set to t in equation (7)_dDeriving the current SOC and v_1,0And v_2,0At the current state of (1), the duration Δ t ═ t_dInner current limit.

Can set t as t_dThe following equation is derived by combining equation (7) and equation (9) at the same time:

$v_t=v_{\mathrm{OC}}-(1-\frac{1}{R_1{C}_1})v_{\mathrm{1,0}}-(1-\frac{1}{R_2{C}_2})v_{\mathrm{2,0}}-(R_0+(\frac{1}{C_1}+\frac{1}{C_2})t_d)i---\left(10\right)$

by setting v as shown in the following equation_t＝v_lbTo calculate the battery discharge current limit i according to equation (10)_dch,limWherein v is_lbIs the battery low voltage limit:

$i_{\mathrm{dch},\max }=[(v_{\mathrm{OC}}-v_{\mathrm{lb}})-(1-\frac{1}{R_1{C}_1})v_{\mathrm{1,0}}-(1-\frac{1}{R_2{C}_2})v_{\mathrm{2,0}}]/(R_0+(\frac{1}{C_1}+\frac{1}{C_2})t_d).---\left(11\right)$

equation (11) is converted into a state feedback form expressed in the following equation:

i_dch,max＝K₀(v_OC-v_lb)-K₁v_1,0-K₂v_2,0(12)

wherein, $K_0=1/(R_0+(\frac{1}{C_1}+\frac{1}{C_2})t_d),---\left(13a\right)$

wherein, $K_1=(1-\frac{1}{R_1{C}_1})/(R_0+(\frac{1}{C_1}+\frac{1}{C_2})t_d),---\left(13b\right)$

wherein, $K_2=(1-\frac{1}{R_2{C}_2})/(R_0+(\frac{1}{C_1}+\frac{1}{C_2})t_d),---\left(13c\right)$

in equations (13a), (13b), and (13c), the feedback gain, which may be estimated by an Extended Kalman Filter (EKF) or other known estimation methods, is expressed as a function of the battery model parameters.

Fig. 8 is a generalized state feedback structure 600 represented in equation (12), where equation (12) is used to estimate battery parameters using feedback gains 604, 606, 608. The generalized state feedback structure 600 may include a model-based observer 602 using a kalman extension filter.

By setting v_t＝v_ubCalculating battery charging current limit i from equation (10)_chg,lim. Wherein v is_ubThe high voltage limit of the battery. By setting v_t＝v_lbCalculating the battery discharge current limit i from equation (10)_dchg,lim. The battery charging current limit 610 using the generalized state feedback gains 604, 606, 608 is represented in the following equation:

i_ch,max＝K₀(v_OC-v_ub)-K₁v_1,0-K₂v_2,0.(14a)

i_dch,max＝K₀(v_OC-v_lb)-K₁v_1,0-K₂v_2,0(14b)

where equation (14b) is shown as an example of the state feedback structure 600 in fig. 8. If an additional RC circuit is used in the battery equivalent circuit model,the same process can be used to obtain the information for the additional state variable v_jAdditional feedback gain K of_j. Thus, the discharge current limit and the charge current limit using two or more RC circuits are represented in the following equations:

i_dch,max＝K₀(v_OC-v_lb)-K₁v_1,0-...-K_nv_n,0,(15a)

i_ch,max＝K₀(v_OC-v_ub)-K₁v_1,0-...-K_nv_n,0.(15b)

the battery management system for calculating the battery model parameters can implement a compact feedback form that flexibly extends the number of state variables and the current limit estimation structure. Once the model parameters are determined to be offline or online, the feedback gain may be calculated. After the feedback gain is calculated, the battery current limit is calculated.

To be shown in the following equation, the feedback gain K₀、K₁And K₂Can be modified to take into account possible noise and errors and to improve the robustness of the estimated current limit:

<math> <mrow> <msub> <mi>K</mi> <mn>0</mn> </msub> <mo>=</mo> <msub> <mi>&alpha;</mi> <mn>0</mn> </msub> <mo>/</mo> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mn>0</mn> </msub> <mo>+</mo> <mrow> <mo>(</mo> <mfrac> <mn>1</mn> <msub> <mi>C</mi> <mn>1</mn> </msub> </mfrac> <mo>+</mo> <mfrac> <mn>1</mn> <msub> <mi>C</mi> <mn>2</mn> </msub> </mfrac> <mo>)</mo> </mrow> <msub> <mi>t</mi> <mi>d</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>16</mn> <mi>a</mi> <mo>)</mo> </mrow> </mrow> </math>

<math> <mrow> <msub> <mi>K</mi> <mn>1</mn> </msub> <mo>=</mo> <msub> <mi>&alpha;</mi> <mn>1</mn> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <mfrac> <mn>1</mn> <mrow> <msub> <mi>R</mi> <mn>1</mn> </msub> <msub> <mi>C</mi> <mn>1</mn> </msub> </mrow> </mfrac> <mo>)</mo> </mrow> <mo>/</mo> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mn>0</mn> </msub> <mo>+</mo> <mrow> <mo>(</mo> <mfrac> <mn>1</mn> <msub> <mi>C</mi> <mn>1</mn> </msub> </mfrac> <mo>+</mo> <mfrac> <mn>1</mn> <msub> <mi>C</mi> <mn>2</mn> </msub> </mfrac> <mo>)</mo> </mrow> <msub> <mi>t</mi> <mi>d</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>16</mn> <mi>b</mi> <mo>)</mo> </mrow> </mrow> </math>

<math> <mrow> <msub> <mi>K</mi> <mn>2</mn> </msub> <mo>=</mo> <msub> <mi>&alpha;</mi> <mn>2</mn> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <mfrac> <mn>1</mn> <mrow> <msub> <mi>R</mi> <mn>2</mn> </msub> <msub> <mi>C</mi> <mn>2</mn> </msub> </mrow> </mfrac> <mo>)</mo> </mrow> <mo>/</mo> <mrow> <mo>(</mo> <msub> <mi>R</mi> <mn>0</mn> </msub> <mo>+</mo> <mrow> <mo>(</mo> <mfrac> <mn>1</mn> <msub> <mi>C</mi> <mn>1</mn> </msub> </mfrac> <mo>+</mo> <mfrac> <mn>1</mn> <msub> <mi>C</mi> <mn>2</mn> </msub> </mfrac> <mo>)</mo> </mrow> <msub> <mi>t</mi> <mi>d</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>16</mn> <mi>c</mi> <mo>)</mo> </mrow> </mrow> </math>

wherein alpha is₀、α₁And alpha₂Are the adjustment parameters of the feedback gain, respectively.

The battery model parameters may be calibrated off-line or estimated in real-time. If real-time model parameter estimation is used, EKF may be used. The EKF used to estimate the model parameters and state variables is formulated by the following process represented in the following formula:

${\hat{x}}_{k|k-1}=f({\hat{x}}_{k-1|k-1},u_{k-1})---\left(17\right)$

wherein,is an augmented state vector, u_k-1Is the input current.

Input current u_kIs passed to the algorithm at a specific operating point to allow the system to predict battery parameters at a time variable. The model parameters are used to predict the voltage response when a constant current is applied over a period of time. Based on the above equation and the EKF known variables, the updated filter equation can now predict the next state of battery power capacity using the following prediction covariance equation:

<math> <mrow> <msub> <mi>F</mi> <mrow> <mi>k</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>=</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>f</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>x</mi> </mrow> </mfrac> <msub> <mo>|</mo> <mrow> <msub> <mover> <mi>x</mi> <mo>^</mo> </mover> <mrow> <mi>k</mi> <mo>-</mo> <mn>1</mn> <mo>|</mo> <mi>k</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>,</mo> <msub> <mi>u</mi> <mrow> <mi>k</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> </mrow> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>18</mn> <mi>a</mi> <mo>)</mo> </mrow> </mrow> </math>

$P_{k|k}=\mathrm{cov}(x_k-{\hat{x}}_{k|k})=E\left((x_k-{\hat{x}}_{k|k}){(x_k-{\hat{x}}_{k|k})}^T\right)---\left(18b\right)$

$P_{k|k-1}=\mathrm{cov}(x_k-{\hat{x}}_{k|k-1})=E\left((x_k-{\hat{x}}_{k|k-1}){(x_k-{\hat{x}}_{k|k-1})}^T\right)---\left(18c\right)$

$P_{k|k-1}=F_{k-1}{P}_{k-1|k-1}{F}_{k-1}^T+Q_k---\left(19\right)$

the new measurement y is calculated using the following equation_kAnd the predicted valueThe difference between:

${\tilde{y}}_k=y_k-h\left({\hat{x}}_{k|k-1}\right)---\left(20\right)$

as shown in the following equation, the following equation is for determining the Kalman gain K_kIntermediate amount of (2):

$S_k=H_k{P}_{k|k-1}{H}_k^T+R_k---\left(21\right)$

the method for determining the Kalman gain K is shown in the following equation_kEquation (c):

$K_k=P_{k|k-1}{H}_k^T{S}_k^{-1}---\left(22\right)$

kalman gain is based onDetermining an update status vector

${\hat{x}}_{k|k}={\hat{x}}_{k|k-1}+K_k{\tilde{y}}_k---\left(23\right)$

The covariance of the state estimation error is shown in the following equation:

P_k|k＝(I-K_kH_k)P_k|k-1(24)

the model parameters are estimated according to equation (24).

The system may calculate the battery instantaneous power capacity during a charging event using the following equation:

P_lim＝|i_chg,max|v_ub(25)

wherein, P_limTo power capacity, v_ubIs the upper limit of the battery voltage, i_minIs the absolute minimum current.

The system may calculate the battery instantaneous power capacity during a discharge event using the following equation:

P_lim＝|i_dch,max|v_lb(26)

wherein, P_limTo power capacity, v_lbIs the lower limit of the battery power voltage, i_maxIs the maximum current.

Fig. 9 is a graph 700 illustrating the response of internal state variables of an equivalent circuit battery model. The graph has an x-axis 702 representing time in seconds and a y-axis 704 representing measured voltage. Respectively passing through the first RC circuit₁706 and a second RC circuit voltage v₂708 capture the different frequency dynamics. Voltage v₁706 may represent a fast dynamic behavior while the voltage v₂708 may indicate slow dynamics. The voltage response may be used to estimate battery current limits, battery power capacity, and other battery performance variables.

Fig. 10 is graphs 800, 801 illustrating current limits using a state feedback structure. The graphs 800, 801 have an x-axis 802 representing time in seconds and a y-axis 804 representing current in amperes. Graph 800 shows the maximum discharge current with and without a state feedback structure, and graph 801 shows the maximum charge current with and without a state feedback structure.

The maximum discharge current plot 800 shows a current limit that may be modeled using a state feedback structure to improve battery parameters. The maximum discharge current with state feedback 808 may have an improved calculation of the current limit compared to the maximum discharge current without state feedback 806.

The maximum charge current plot 801 shows current limits that may be modeled using state feedback structures to improve battery parameters. The maximum charging current with state feedback 812 may have an improved calculation of the current limit compared to the maximum charging current without state feedback 810.

Fig. 11 is a flow diagram of an algorithm of a method 900 of estimating battery current limits and battery power limits in a battery management system, according to an embodiment. According to one or more embodiments, the method 900 is implemented using software code contained within a vehicle control module. In other embodiments, the method 900 is implemented in other vehicle controllers or distributed among multiple vehicle controllers.

Referring again to fig. 11, throughout the discussion of method 900, reference is made to the vehicle and its components shown in fig. 1, 3, 5, and 8 to facilitate an understanding of various aspects of the present disclosure. In a hybrid electric vehicle, the method of using a state feedback structure to estimate battery performance variables to determine current limits and power limits may be implemented by a computer algorithm, machine executable code, or software instructions programmed into a suitable programmable logic device of the vehicle (e.g., a vehicle control module, a hybrid control module, another controller in communication with a vehicle computing system, or a combination of the foregoing). Although the various steps shown in the flow diagrams appear to occur in a chronological order, at least some of the steps may occur in a different order, and some steps may or may not be performed simultaneously.

At step 902, the vehicle computing system may begin powering one or more modules during a key-on event that allows the vehicle to be powered on. At step 904, the powering of the one or more modules may cause a variable associated with the battery management system to be initialized prior to enabling one or more algorithms for controlling the battery.

The initialized parameter may be a value stored in the last key-off event or a predetermined value. The parameters should be initialized before the algorithm is enabled at the key-on event. For example, a battery management method may initialize several variables (including, but not limited to, battery terminal voltage, current limits, and/or other battery-related parameters).

At 906, the system may measure the battery voltage output and current input in real time using several types of sensors. Once the system receives the battery voltage response and the battery current measurement, the system may process the received signals to calculate a battery state variable, wherein the battery state variable is represented by a voltage response based on fast and slow dynamics of the battery.

At step 908, two or more RC circuits in the equivalent circuit mode may be used to measure model parameter estimates for fast dynamic voltage responses and model parameter estimates for slow dynamic voltage responses. EKF can be used for model parameter estimation. The EKF-based model parameter estimates are obtained using equations (17) - (24). Other online parameter estimation methods may be used if the model parameters can be determined in real time. If an online parameter estimation method is not used in the battery management system, an offline calibrated model parameter map may be used.

At step 910, the system may estimate a state variable. The state variable comprising the open-circuit voltage v of the battery_OC216. Voltage v across an RC circuit consisting of fast dynamic characteristic voltage response₁210. RC circuit composed of slow dynamic characteristic voltage responseVoltage v across₂408. Open circuit voltage v_OC216 may be estimated based on battery state of charge, which may be calculated by current integration or other algorithms.

In another embodiment, step 908 and step 910 may be combined into a single step performed by the system. For example, the estimation process may include the battery model parameters and the battery state variables in one estimation structure (referred to as "parameter state co-estimation"). In this embodiment, different time scales in parameter changes and state changes may result in some degradation in estimation performance, but the estimation structure may be a simpler model calculated by the system. However, separating the state variable estimation process from the estimation of the model parameters may allow the system to improve the estimation accuracy for each state variable and each model parameter.

In step 912, the feedback gain may be calculated by equation (13a), equation (13b), and equation (13 c). When possible measurement and process noise and errors need to be suppressed, the feedback gain can be calculated by equation (16a), equation (16b), and equation (16c) to improve the robustness of estimating the current limit.

In step 914, the system may calculate the current limit using a state feedback structure for fast dynamics, slow dynamics, and battery open circuit voltage as shown in equations (14a) and (14 b). If an additional RC circuit is used in the battery equivalent circuit model, the discharge current limit and the charge current limit using two or more RC circuits are expressed in accordance with equations (15a) and (15 b).

In step 916, the system may calculate the power limit using equations (25) and (26). The calculated power limit may be used to determine a battery current command from the battery controller to the battery pack.

At step 918, if the system detects a key-off event, the system may end one or more algorithms for managing the battery pack and/or one or more battery cells. At step 920, the vehicle computing system may have a vehicle key-off mode for allowing the system to store one or more parameters in non-volatile memory so that these parameters may be used by the system for the next key-on event.

While exemplary embodiments are described above, these embodiments are not intended to describe all possible forms of the invention. Rather, the words used in the specification are words of description rather than limitation, and it is understood that various changes may be made without departing from the spirit and scope of the invention. Furthermore, features of various implementing embodiments may be combined to form further embodiments of the invention.

Claims (7)

1. A battery management system, comprising:

a battery pack;

at least one controller configured to: operating the battery pack within a current limit derived from a first state variable value and a second state variable value and a feedback gain, wherein the first state variable value is based on battery pack terminal voltage data and battery pack current data and has a frequency component greater than a threshold value, and the second state variable value is based on battery pack terminal voltage data and battery pack current data and has a frequency component less than the threshold value.

2. The system of claim 1, wherein the feedback gain is based on an impedance parameter defining an equivalent circuit model of the battery pack.

3. The system of claim 2, wherein the equivalent circuit model comprises two or more RC circuits.

4. The system of claim 1, wherein the current limit comprises a discharge current limit and a charge current limit.

5. The system of claim 1, wherein the at least one controller is further configured to: operating the battery pack during a charging event or during a discharging event within power limits including an instantaneous power capacity derived from the battery pack terminal voltage data and the battery pack current data.

6. The system of claim 1, wherein the at least one controller is further configured to: operating the battery pack during a charging event or during a discharging event within power limits including a sustained power capacity derived from the battery pack terminal voltage data and the battery pack current data.

7. The system of claim 1, wherein the at least one controller is further configured to: operating the battery pack within a power limit having a predefined duration during a charging event or during a discharging event.